ABSTRACT Pemetrexed is a multitargeted antifolate recently approved for second line therapy of non-small cell lung cancer (NSCLC). The most significant interaction at clinically achievable concentrations is with thymidylate synthase (TYMS), although the drug is known to target at least two other folate dependent enzymes. The goal of this proposal is to identify additional genetic variants that influence pemetrexed-associated toxicity and response by employing a genome-wide, comprehensive approach to test the hypothesis that genetic polymorphisms significantly influence susceptibility to toxicities and response associated with pemetrexed. To this end, lymphoblastoid cell lines, derived from large reference Centre d' Etude du Polymorphisme Humain pedigrees and HapMap (Caucasian, Yoruban, Asian) samples, will be phenotyped for pemetrexed cytotoxicity and apoptosis. Genotypic information is available in the public domain enabling us to apply familial genetic strategies, including linkage analysis, to examine regions in the genome harboring the genetic variants responsible for susceptibility to pemetrexed-induced cytotoxicity and apoptosis. In addition, inter-individual and inter-ethnic variation in pemetrexed-induced cytotoxicity will be compared using the International HapMap samples from populations of African, European and Asian descent. Exon expression array generated in our laboratory will be incorporated to identify genetic variants acting through their effect on baseline gene expression that associate with sensitivity to pemetrexed. Genetic variants identified through this whole genome approach will be validated in a separate sample set of LCLs and by quantitative PCR and siRNA. We will evaluate both novel SNPs and known candidate variants in TYMS, GARFT, DHFR to determine genotypes associated with response or toxicity in an intergroup (NCCTG, CALGB, ECOG, SWOG, NCI-C) study of pemetrexed compared to erlotinib as second line therapy for NSCLC. Our long-term goal is to identify a pharmacogenetic signature that will predict patients with a decreased chance for response or an increased risk for adverse reactions to pemetrexed. Specific aims are: 1. To determine the genetic contribution of susceptibility to the cytotoxic and apoptotic effects of pemetrexed. 2. To examine baseline and time dependent patterns of expression following pemetrexed treatment. 3. Identify SNPs, through data available from the International HapMap project, associated with both pemetrexed cytotoxicity and gene expression in cell lines derived from Europeans (CEU), Africans (YRI) and Asians (CHB, JPT). 4. To determine if a common polymorphism in TYMS, GARFT, and DHFR, as well as candidate polymorphisms identified in specific aims 1- 3 correlate with response and/or toxicity in advanced NSCLC patients treated with pemetrexed in a multicenter trial.